L Graftieaux撰写了一篇关于数学的问题“Combining PIV, POD and vortex identification algorithms for the study of unsteady turbulent swirling flows”https://iopscience.iop.org/article/10.1088/0957-0233/12/9/307/pdf

Underlying math presented by L Graftieaux "Combining PIV, POD and vortex identification algorithms for the study of unsteady turbulent swirling flows" https://iopscience.iop.org/article/10.1088/0957-0233/12/9/307/pdf 

由于Sebastian的代码运行很慢，因为它使用了4层循环嵌套。

Sebastian's code is rather slow, though, because it uses 4 nested for loops. 

通过在Matlab中实现将Graftieaux论文中描述的操作重新排列为两个卷积，可以更有效地生成gamma1映射。

Producing the gamma1 map can be done more efficiently in Matlab by realizing that the operations described in the Graftieaux's paper can be rearranged into two convolutions. 

它还通过拟合一条样条曲线找到最大的gamma 1，因此不受矢量场的网格分辨率限制。

It also find the maximum gamma 1 by fitting a spline curve, therefore not being limited by the grid resolution of the vector field.

缺点是它只能找到一个涡核，这在一些形成多个涡流的应用中可能是不可行的。

The drawback is that it'll find only one vortex core, which might not be desired in some applications where multiple vortices form. 

但是，可以使用第二个输出（G1）执行峰值/阈值查找来搜索多个涡核。

One can, however, use the second output (G1) to perform their own peak finding/thresholding to find multiple vortex cores.

更多精彩文章请关注公众号：